Thanks Jim for this initial email. It would be indeed good to better understand the needs and required changes, to motivate the effort we put in the Lyo project.
I can complement the list below with the following ideas that we (members from KTH) are willing to contribute with: 1. Documentation - http://open-services.net/resources/ is a very good source of information on OSLC, and Lyo. The threshold to learn the technologies seems still to be high for new comers. What kind of resources are we missing to make this easier? 2. Continue to extend the graphical modelling tool and accompanying generator. Potentially, we can further automate the generation process for specific families of tools, such as SQL-based or EMF-based tools. We can prioritize this depending on interest and needs. 3. Data validation - provide support to validate that produced/consumed resources actually match the define resource shapes. 4. Support for the storage and handling of resources within the OSLC server. This is driven by the need to (1) improve the performance of OSLC servers, whereby multiple REST requests need not necessarily lead to multiple requests to the source tool. (2) provide more capable SPARQL query support. regards ______________________________ Jad El-khoury, PhD KTH Royal Institute of Technology School of Industrial Engineering and Management, Mechatronics Division Brinellvägen 83, SE-100 44 Stockholm, Sweden Phone: +46(0)8 790 6877 Mobile: +46(0)70 773 93 45 [email protected]<mailto:[email protected]>, www.kth.se<http://www.kth.se> From: [email protected] [mailto:[email protected]] On Behalf Of Jim Amsden Sent: 03 January 2017 18:34 To: [email protected]; [email protected] Subject: [lyo-dev] What's next for Eclipse/Lyo? Happy New Year to the Eclipse/Lyo and open-services.net communities. As we start 2017, perhaps we should reflect on what's next for Eclipse/Lyo, especially given changes in the dev team, the OASIS OSLC Core 3.0 specification, and integration in the marketplace. Certainly lifecycle collaboration tools have reached a level of maturity over the last few years, and OSLC 2.0 has played a central and largely successful role in enabling loosely coupled integration between lifecycle tools. Jad has made some great progress in the Lyo code generators to provide a simple means of designing and implementing tool chains supporting integrated OSLC domains. Where we go next depends a lot on what you, the Lyo community needs. There is always a gap and tension between the demand and supply side views of any offering. But open-source is essentially demand driven. So to better understand your needs, let's start an email discussion to better determine who's using Eclipse/Lyo, and what changes they'd like to see. I'll try to collate the results of the email into a Doodle poll to help us determine priorities. Then we can use this to guide planning this year. To get things started, here's a few things we might consider doing. 1. Update Eclipse/Lyo dependencies - some of the dependencies are pretty old and might present a barrier to adopting Lyo in more up-to-date environments. I have made these changes on a local branch, but have not yet done the CQs in order to commit them and create an updated release 2. Update Eclipse/Lyo distribution packages - Lyo consists of a lot of Git repositories and eclipse projects. This makes it hard for someone to know what they should actually use to meet their needs. We might organize the eclipse projects into discrete packages that can be built and released independently and would be easier for users to understand and use. 3. Use Maven Central for Eclipse/Lyo distribution packages - Currently different releases of Eclipse/Lyo are distributed as Jar files accessed from eclipse.org. There is also a Maven repository at https://wiki.eclipse.org/Hudson/Maven. Lyo might benefit from being able to access the Lyo packages and versions through Maven Central. 4. Create a Lyo 3.0 version compatible with OASIS OSLC Core 3.0 - Its taken a lot longer to get the Core 3.0 specification completed and standardized. The OASIS OSLC Core TC is about to vote for Committee Specification which is the last step before OASIS Standard. This means that Core 3.0 has been through public review and is stable enough to implement. To become an OASIS Standard, OSLC Core 3.0 needs three "statements of use" or implementations. Lyo could be one or more of these. Since Core 3.0 is backward compatible with Core 2.0, this could be done by extending Lyo without breaking existing OSLC 2.0 clients (consumers) and servers (providers). 5. Extend Lyo to support OSLC Configuration Management 1.0 specification - this would allow Lyo to support development of clients and servers that support configuration management. 6. Infrastructure Simplification - look at the use of Apache Wink and other technologies, and the extensive use of Java annotations and see if there are some runtime platform simplifications we could do to expand the applicability of OSLC and Lyo 7. Development of OSLC clients and servers in other environments such as JavaScript/Node, .Net, MacOS/Swift, etc. See OSLC4JS for example at http://oslc.github.io/developing-oslc-applications/oslc-open-source-node-projects.html Jim Amsden, Senior Technical Staff Member OSLC and Linked Lifecycle Data 919-525-6575
